Subscriber access provided by Eastern Michigan University | Bruce T. Halle Library
Letter
Think Negative: Finding the best ESI/MS mode for your analyte Piia Liigand, Karl Kaupmees, Kristjan Haav, Jaanus Liigand, Ivo Leito, Marion Girod, Rodolphe Antoine, and Anneli Kruve Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 10 May 2017 Downloaded from http://pubs.acs.org on May 10, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Think Negative: Finding the best ESI/MS mode for your analyte Piia Liigand*,† Karl Kaupmees,† Kristjan Haav,† Jaanus Liigand,† Ivo Leito,† Marion Girod,‡ Rodolphe Antoine, ¶ Anneli Kruve†,§ †
Institute of Chemistry, University of Tartu, Ravila 14a, 50411, Tartu, Estonia
‡
Univ Lyon, CNRS, Université Claude Bernard Lyon 1, Ens de Lyon, Institut des Sciences Analytiques, UMR 5280, 5 rue de la Doua, F69100 Villeurbanne, France ¶ Univ
Lyon, Université Claude Bernard Lyon 1, CNRS, Institut Lumière Matière, UMR 5306 F-69622, LYON, France
§Schulich
Faculty of Chemistry, Technion - Israel Institute of Technology, Technion City, Haifa 3200008, Israel
ABSTRACT: For the first time the electrospray ionization efficiency (IE) scales in positive and negative mode are united into a single system enabling direct comparison of IE values across ionization modes. This is made possible by the use of a reference compound that ionizes to a similar extent in both positive and negative modes. Thus, choosing the optimal (i.e. most sensitive) ionization conditions for a given set of analytes is enabled. Ionization efficiencies of 33 compounds ionizing in both modes demonstrate that contrary to general practice, negative mode allows better sensitivity for 46% of such compounds whereas the positive mode is preferred for only 18%, and for 36% the results for both modes are comparable.
Mass spectrometry (MS) with electrospray ionization (ESI1) source is a key technique in various research fields, ranging from food and environmental analysis to metabolomics and proteomics.2 However, despite its widespread application, current understanding of the ESI process is still limited. 3,4 Ionization efficiency (IE), the amount of ions generated from a specific compound in the ionization source, may vary from compound to compound by more than six orders of magnitude 3,5–10. Further, IE in ESI is highly dependent on the solvent3,5,11–14 and ionization mode 3. Higher IEs enable better sensitivity and lower detection limits. Therefore, finding the setup whereby the analyte of interest has the highest (or high enough) IE allows lower limits of detections. If ESI/MS is coupled with liquid chromatography (LC), ESI+ is generally preferred as more compounds are expected to ionize in this mode. 3,5 However, the major advantage of negative ion mode (ESI-) is the lower background noise.3,5 Currently, there is a paucity of research and guidelines available5 on which mode to choose when a compound ionizes in both modes. In the interpretation of ESI process one important factor is the solvent composition, which is usually described in terms of initial composition, since the actual composition in the plume is difficult to measure. However, it has been shown that solvent pH, 15–18 organic solvent content, 19–23 and droplet size 19,22,24,25 change along the plume. In a recent study,26 we have shown that changes in pH, organic content and droplet size are similar for ESI+ and ESI-. This allows comparison of ESI+ and ESI- modes without the need to account for major solvent property changes between the two modes.
Absolute IEs are impractical to determine. Therefore, relative ionization efficiencies (IE of the analyte relative to an anchor substance) are commonly used. In both ESI+ and ESI-, extensive IE scales have been constructed in 80/20 acetonitrile (MeCN)/0.1% formic acid 6 and 80/20 MeCN/0.1% ammonia solution, 7 respectively. All values on both scales are given relative to another compound, for ESI+ relative to methyl benzoate (MB) and in ESI- relative to benzoic acid (BA). However, the ionization of methyl benzoate in ESI+ and benzoic acid in ESI- cannot be compared directly.
Figure 1. Process of unifying the different IE scales measured in different solvents and in ESI+ and ESI-. In this study, we aimed for the first time to make ESI+ 6 and ESI-7 IE scales quantitatively comparable in order to give further insight into the differences and similarities of the ESI modes and enable selection of the most suitable mode for any given compound. This is accomplished (as depicted in Figure 1) by (1) characterizing and comparing the ESI plume for ESI+ and ESI-, (2) using a reference compound with both acidic and basic moieties with similar
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
pKa values, allowing ionization to similar extent in both modes from the same solution, and (3) measuring the links from ESI+ and ESI- anchoring compounds to the reference compound (in respective solvents) and unifying ESI+ and ESI- scales.
EXPERIMENTAL SECTION Ionization efficiencies were measured with Agilent 6496 Triple Quadrupole mass spectrometer with Agilent Jet Stream source (Agilent Technologies, Santa Clara, CA, USA) and Agilent XCT ion trap spectrometer (See Supporting Information (SI) for further details). Relative logIE values were obtained from the slope of the corresponding calibration graphs of compound and anchor: slope(compound) (Eq. 1) log IE (compound) = log slope(anchor) where the anchor compound is BA (Reakhim, Russia) for ESI-, in MeCN/0.1% ammonia solution 80/20 or tetraethylammonium perchlorate (Fluka, Buchs, Switzerland) for ESI+, in MeCN/0.1% HCOOH solution 80/20. The difference between logIEESI- values in two different solvents were obtained from MS signal intensities for BA solution (concentration in both solutions of approximately 60 μmol/L) as according to Eq. 2 and averaged over three days: solvent(1) log IEBA = log
solvent(1) RBA solvent(2) RBA
solvent(2) CBA solvent(1) CBA
(Eq. 2)
Page 2 of 5
extent of ionization in the ESI source. Therefore, it was necessary that basic and acidic groups of the anchoring compound would have similar ionization degrees in solution. Since the physicochemical changes in the ESI plume are similar in ESI+ and ESI-, the ionization in both modes occurs from fairly similar solutions. 26 This significantly simplifies choosing a suitable solvent system for anchoring measurements. Finding a suitable reference compound. We tested several compounds (see SI) and observed the closest degrees of ionization of the acidic and basic groups for TA in 80/20 MeCN/pH 4.00 (v/v) solution. The degree of ionization for the acidic group was 0.38 and for the basic group, 0.19. Since the pH in the plume region close to the MS entrance is somewhat lower (ca 0.8 for ESI+ and 0.1 units for ESI-26) than the initial pH, the degrees of ionization are in fact even more similar and the ionization of TA in ESI+ and ESI- will also be similar. Here, we focus on the ionization of a specific group and not on the ionization of the compound as such, as some of its molecules may be in solution as zwitterions. However, the simultaneous ionization of two groups is expected to affect ionization in both modes in the same way, as statistically the same fraction of molecules is present in the droplets as zwitterions in both modes. Linking the scales. The difference of anchor values ESI− (ΔESI+vs ) of ESI+ and ESI- can be derived (Figure 2): anchors vsESI c c pH4 ESI log IE ESI ,0.1%formi log IE ESI ,0.1%formi anchors Et 4 N MB
where R is the response of the compound and C is concentration. The degree of ionization of the reference compound, trans-3(3-pyridyl)acrylic acid (TA) (Aldrich, St. Louis, USA), was measured using 1H-NMR and UV-Vis spectroscopy (see details of apparatus in SI). For NMR, the chemical shift of the protons was used to calculate the degrees of ionization for the reference compound in the solvents (based on the calibration with solutions with known degrees on ionization, see Figure S1 and Table S2). Similar approach was used for UV-Vis. In addition, 33 compounds (amino acids, substituted benzoic acids and phenols, oligopeptides, and polyfunctional aromatic compounds) were chosen for comparison of IE values in ESI+ and ESI-. See Table S1 for more details. Weighted average positive and negative sigma (WAPS27 and WANS28, showing the degree of charge delocalization) values were calculated with the COSMO-RS29 method. pKa values were calculated with ACE and JChem acidity and basicity calculator30 (see SI).
Et 4 N
ESI , pH4 ESI ,0.1%ammon ia pH4 log IE ESI , pH4 log IETA BA log IE BA TA Et 4 N
Where superscript denotes the ESI mode and media that the IE was measured in; subscript denotes the compounds that were measured. ESI+,pH 4 The log 𝐼𝐸TA−Et N+ of TA was measured to be -1.11 in 4
ESI−,pH 4
ESI+ (anchor substance Et 4N+) and log 𝐼𝐸TA−BA was 0.73 in ESI- (anchor substance BA), both in 80/20 MeCN/pH 4.00 (v/v) solution. However, logIE scales in ESI+ and ESI- have been compiled in different solutions and to make the scales comparable, an anchor substance, BA in 80/20 MeCN/pH 4.00 (v/v) solution and 80/20 MeCN/0.1% ammonia (v/v) solution (ESI-), was infused. ESI−,0.1%ammonia−pH 4
A difference (log 𝐼𝐸BA ) of 0.90 logarithmic units was observed, meaning that the IE of BA is 0.90 logarithmic units lower in 80/20 MeCN/pH 4.00. In ESI+ the Et4N+ in 80/20 MeCN/0.1% formic acid (v/v) ESI+,0.1%formic solution has an IE of (log 𝐼𝐸Et ) 3.95 relative to + 4 N −MB
methyl benzoate.6 It has been observed previously that the IE of Et4N+ is not affected by changes in pH ESI+,0.1%formic−pH 4 + 4N
(log 𝐼𝐸Et
RESULTS AND DISCUSSION Characterizing the ESI plume. To make ESI+ and ESIscales comparable, it is important to find a system where logIEESI+ and logIEESI- would become similar (steps (2) and (3) on Figure 1). It has previously been observed7 that among other things, the ionization of compounds in the solvent and affinity towards droplet surface determines the
(Eq 3)
= 0).23
Unifying IE scales. Substituting these values into Eq 3 gives a difference of 3.95 – 1.11 – 0.73 + 0.90 = 3.01 logarithmic units between the IEs of anchors between ESI+ and ESI- (Figure 2). Knowing the difference in anchors allows direct comparison in logIE values between ESI+ and ESI- mode.
ACS Paragon Plus Environment
Page 3 of 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Figure 2. Comparison of logIE scales compiled in ESI+ (logIEESI+) and ESI- (logIEESI-). To further verify this difference, the logIE prediction models were compared (Equations 4 and 5 in Figure ). We have previously developed a model (Eq. 5) for predicting IEs in ESI-7 (logIEESI-) based on the degree of ionization and the charge delocalization in anions (WAPS).27 In order to make the IE prediction models comparable, we have modified the model for logIE values on the ESI+ scale 6 to contain a WANS parameter that describes charge delocalization in cations (see Table S3, Figure S3),28 and obtained Eq. 4 for finding logIEESI+. The ability to predict IEs is similar (standard residual error sRE = 0.87 logIE units) to the previous version6 of the model (sRE = 0.86 logIE units). The intercept of such models carries information about the logIE (relative to the anchoring compound) of a hypothetical compound which is neutral (degree of ionization, α = 0) but has infinitely delocalized charge (WANS or WAPS = 0). Such compounds should ionize to the same extent in ESI+ and ESI-. Therefore, the numerical value of intercept carries information about the anchoring compound and the difference in intercepts allows comparing the anchoring values. The difference in the intercepts (2.97) in the logIE prediction models for ESI+ and ESI- (Figure 2) was statistically insignificantly ESI− different (t-test on 95% confidence level) from ∆ESI+vs anchors (3.01). This implies that the difference is not accidental and has been correctly assigned. Comparison of the two ionization modes. We applied this knowledge to a set of 33 compounds (Table S1), which ionize in both ESI+ (a range of 3.5 logIE units) and ESI- (a range of 3 logIE units) and found the difference between logIEESI+ and logIEESI- by taking into account the difference in the anchors (Figure 3). The difference in logIE values (logIEESI+ - logIEESI-) was compared against a constant of 0.3 logarithmic units, which refers to a two times difference in signal scale. We find that a two-fold increase is of significance for practitioners and is also statistically significant (based on repeatability of the measurements). Out of the investigated compounds, which all can be ionized in both modes, for six compounds ESI+ is preferred (logIE difference > 0.3 log units), and for 15 compounds ESI- is preferred (logIE difference < -0.3). Twelve compounds ionize to a similar degree in ESI+ and ESI-. Thus, in approximately 46% of cases a compound is better ionized in ESI-, and indifferent for 36%.
IE is considerably enhanced (up to 100 times) in ESIfor compounds with low logIEESI+. More precisely, ESI- is preferred by compounds that are only oxygen bases (e.g. mostly carboxylic acids). Compounds being simultaneously both oxygen (carboxylic acid) and nitrogen bases fall into all three categories. For small peptides and amino acids, the differences between ESI+ and ESI- are mostly very small and either of the modes could be used. One of the exceptions is histidine, which contains a basic side chain that may account for a strong preference towards ESI+ mode (0.72 logIE units, 5.3 times). Aspartic acid and glutamic acid, amino acids with acidic side chain, show weak preference towards ESI-. When interpreting these results, it is of course important to keep in mind that all the investigated compounds have acidic groups in their structure. From previous studies6,7 we know that IE in both modes is best described by degree of ionization in solution (α) and charge delocalization (WAPS or WANS). For most compounds, the ionization degree in solution is sufficient to explain the ionization mode preference, including examples described above. For example, compounds that are oxygen bases (such as carboxylic acids) tend to be very weak bases but are at the same time medium strength acids. Therefore, the formation of anions is preferred for these compounds. For amino acids both amino and carboxylic acid group are expected to be charged, and therefore only small differences between ESI+ and ESI- are expected.
Figure 3. Difference in logIE values for compounds ionizing in both ESI+ and ESI- between the two modes. Exceptions to this rule are the amino acids tyrosine and methionine. They are both fully ionized in ESI+ and ESIsolvent; however, tyrosine is better ionized in ESI- and methionine is better ionized in ESI+. To explain this phenomenon, we need to look at the ionization process in more detail. There are two main requirements for a compound to become ionized: (1) it needs to become charged, and (2) it needs to partition to the surface of ESI droplet to be ejected to the gas phase. Previously differences in anion and cation surface affinities have been studied for protons and hydroxide ions 31 and significant differences have been observed. These differences are largely determined by ion-solvent interactions, which are different for cations and anions of the same analyte, arising from differences in charge delocalization, stereochemistry and solvent properties. In the case of methionine, the charge in the cation is significantly better delocalized than in the anion, based on the comparison of WAPS and WANS (6.1 and 5.3, respectively, lower value means better charge
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
delocalization). For tyrosine, the charge is slightly better delocalized in the anion (6.0 and 6.3 respectively). Therefore, the preferred ionization mode for tyrosine is opposite to methionine due to opposite charge delocalization of cation and anion. Additionally, tyrosine has an ionizable side-chain in ESI-.32
(6) (7) (8)
(9)
CONCLUSIONS ESI+ and ESI- logIE scales were made comparable, thereby for the first time allowing comparison of ionization efficiency values obtained in both modes. Until now ESI+ has generally been preferred amongst practitioners, however, it can be seen that in many cases, ESI- is the better option owing to its improved sensitivity (ionization efficiency) and potential for lower detection limits. Whilst the solution phase ionization degree mainly explains the ionization mode preference, solvent-induced charge solvation of molecules in solution seems to be a key factor driving the ionization efficiency of molecules in the ESI process.
(10) (11) (12) (13) (14) (15) (16) (17)
ASSOCIATED CONTENT (18)
Supporting Information Detailed materials and methods other additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
(19)
AUTHOR INFORMATION (20)
Corresponding Author * E-mail:
[email protected] (21)
Author Contributions P.L. and K.K. performed UV-Vis spectroscopy measurements, P.L., K.H., K.K. and A.K. performed NMR measurements. P.L., J.L. and A.K. performed ionization efficiency measurements P.L. wrote main part of the text, K.K., I.L., M.G., R.A. and A.K. helped with preparing the manuscript. All authors have given approval to the final version of the manuscript.
(22) (23) (24)
Notes The authors declare no competing financial interest.
(25)
ACKNOWLODGEMENT
(26)
This work was supported by Personal Research Funding Project 34 from the Estonian Research Council.
(27)
REFERENCES (1) (2) (3) (4) (5)
Fenn, J. B. Angew. Chem. Int. Ed. 2003, 42 (33), 3871– 3894. Doerr, A.; Finkelstein, J.; Jarchum, I.; Goodman, C.; Dekker, B. Nat. Methods 2015, 12, S1–S45. Cech, N. B.; Enke, C. G. Mass Spectrom. Rev. 2001, 20 (6), 362–387. Kostiainen, R.; Kauppila, T. J. J. Chromatogr. A 2009, 1216 (4), 685–699. Cole, R. B. Electrospray and MALDI Mass Spectrometry: fundamentals, instrumentation, practicalities, and biological applications.; Wiley: Hoboken, 2010.
(28) (29)
(30) (31)
(32)
Page 4 of 5
Oss, M.; Kruve, A.; Herodes, K.; Leito, I. Anal. Chem. 2010, 82 (7), 2865–2872. Kruve, A.; Kaupmees, K.; Liigand, J.; Leito, I. Anal. Chem. 2014, 86 (10), 4822–4830. Raji, M. A.; Fryčák, P.; Temiyasathit, C.; Kim, S. B.; Mavromaras, G.; Ahn, J.-M.; Schug, K. A. Rapid Commun. Mass Spectrom. 2009, 23 (14), 2221–2232. Ehrmann, B. M.; Henriksen, T.; Cech, N. B. J. Am. Soc. Mass Spectrom. 2008, 19 (5), 719–728. Henriksen, T.; Juhler, R. K.; Svensmark, B.; Cech, N. B. J. Am. Soc. Mass Spectrom. 2005, 16 (4), 446–455. Zhou, S.; Hamburger, M. Rapid Commun. Mass Spectrom. 1995, 9 (15), 1516–1521. Zhou, S.; Cook, K. D. J. Am. Soc. Mass Spectrom. 2000, 11 (11), 961–966. Kruve, A. J. Mass Spectrom. 2016, 51 (8), 596–601. Huffman, B. A.; Poltash, M. L.; Hughey, C. A. Anal. Chem. 2012, 84 (22), 9942–9950. Zhou, S.; Edwards, A. G.; Cook, K. D.; Van Berkel, G. J. Anal. Chem. 1999, 71 (4), 769–776. Zhou, S.; Prebyl, B. S.; Cook, K. D. Anal. Chem. 2002, 74 (19), 4885–4888. Girod, M.; Dagany, X.; Antoine, R.; Dugourd, P. Int. J. Mass Spectrom. 2011, 308 (1), 41–48. Girod, M.; Antoine, R.; Dugourd, P.; Love, C.; Mordehai, A.; Stafford, G. J. Am. Soc. Mass Spectrom. 2012, 23 (7), 1221–1231. Girod, M.; Dagany, X.; Boutou, V.; Broyer, M.; Antoine, R.; Dugourd, P.; Mordehai, A.; Love, C.; Werlich, M.; Fjeldsted, J.; Stafford, G. Phys. Chem. Chem. Phys. 2012, 14 (26), 9389–9396. Zhou, S.; Cook, K. D. Anal. Chem. 2000, 72 (5), 963– 969. Wang, R.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2010, 21 (3), 378–385. Hopkins, R. J.; Reid, J. P. J. Phys. Chem. A 2005, 109 (35), 7923–7931. Liigand, J.; Kruve, A.; Leito, I.; Girod, M.; Antoine, R. J. Am. Soc. Mass Spectrom. 2014, 25 (11), 1853–1861. Wortmann, A.; Kistler-Momotova, A.; Zenobi, R.; Heine, M. C.; Wilhelm, O.; Pratsinis, S. E. J. Am. Soc. Mass Spectrom. 2007, 18 (3), 385–393. Soleilhac, A.; Dagany, X.; Dugourd, P.; Girod, M.; Antoine, R. Anal. Chem. 2015, 87 (16), 8210–8217. Liigand, P.; Heering (Suu), A.; Kaupmees, K.; Leito, I.; Girod, M.; Antoine, R.; Kruve, A. Has Been Submitt. J. Am. Soc. Mass Spectrom. Kaupmees, K.; Kaljurand, I.; Leito, I. J. Phys. Chem. A 2010, 114 (43), 11788–11793. Kaupmees, K.; Kaljurand, I.; Leito, I. J. Solut. Chem. 2014, 43 (7), 1270–1281. Klamt, A. COSMO-RS From Quantum Chemistry to Fluid Phase Thermodynamics and Drug Design.; Elsevier Science Ltd: Amsterdam, 2005. ACE and JChem acidity and basicity calculator https://epoch.uky.edu/ace/public/pKa.jsp. Tse, Y.-L. S.; Chen, C.; Lindberg, G. E.; Kumar, R.; Voth, G. A. J. Am. Chem. Soc. 2015, 137 (39), 12610– 12616. Tian, Z.; Kass, S. R. J. Am. Chem. Soc. 2008, 130 (33), 10842–10843.
ACS Paragon Plus Environment
Page 5 of 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
For table of contents only
5
ACS Paragon Plus Environment
